1. Technical Field
The present invention relates generally to an improved distributed storage system and in particular to a method and a system for configuring a storage area network. Still more particularly, the present invention provides a method and a system for automating the configuration of a storage area network.
2. Description of the Related Art
The primary model of computing has been the mainframe computer with satellite terminals, but this model is being replaced with a client-server model of computing. The client-server model is distributed and may involve a plurality of servers and a plurality of clients, each a powerful workstation in its own right.
Early client-server models inherited from the mainframe model the concept of dedicated storage for each server. There are several major problems associated with using dedicated storage in this environment. Every time a client or a server needs to access information that is not available locally, the computer must issue a request for data to a peer machine. It takes time to process these requests and, if there is a heavy volume of non-local access, a bottleneck may result. In addition, one or more servers may have requirements for data storage that exceed their capacities while other servers may only be using a fraction of their available data storage.
A Storage Area Network (SAN) is a relatively new innovation for the client-server model based on a data-centric infrastructure with a common storage pool. A SAN works well provided there are persistent communications links. Distributed networks, such as the Internet, are constructed with persistence in mind. In other words, the loss of one or several communications links will not shut down the network and an alternative path through the network can almost always be found.
One or more configuration files describing the details of the SAN need to be maintained at each server in the network. A network may contain tens or hundreds of servers and servers may be added or removed on a continual basis. Each configuration file on every server connected to the network must be changed as the machines on the SAN change. It is far too tedious and error prone to manually perform reconfiguration on a server-by-server basis. A SAN may include a variety of operating systems, such as, for example, Solaris(trademark), a version of UNIX used by computers from Sun Microsystems, AIX(trademark), a version of UNIX used by computers from International Business Machines, and Windows NT(trademark), an operating system from Microsoft that is used by a variety of workstations. Transferring data between different operating systems must be done very carefully since the file system formats are not compatible. Transferring data in the wrong format could result in loss of the data and damage to the file system. It only takes a single incorrect entry of the operating system type to cause this problem to occur. If data is being entered and maintained manually on a server that is part of the SAN, this type of error is very probable. Therefore, it would be advantageous to have a master configuration file to represent the complete configuration of the SAN whereby changes to the one master file would be reflected uniformly everywhere on the SAN.
The present invention provides a method and system for the development and maintenance of a single configuration file for a storage area network combined with an xe2x80x9cintelligentxe2x80x9d script which places server configuration files into the appropriate directories for servers that are part of a storage area network. On a server by server basis, each server is configured in accordance with the particular operating system the server itself is using. Once this is accomplished, each server added or removed from the storage area network will be processed in a similar way without the need for reconfiguration of the storage area network as a whole.